Skip to content

Latest commit

 

History

History
83 lines (72 loc) · 3.37 KB

README.md

File metadata and controls

83 lines (72 loc) · 3.37 KB
Raw

Openmmlab-AI-Camp-2th 作业3

训练

  • mmdetection
  • 配置文件：rtmdet_tiny_1xb12-balloon.py
  • 训练完成的模型文件：
    链接：链接：https://pan.baidu.com/s/1uU67QImcqT2z6yl100Hq6g?pwd=huv1
  • 模型在测试集上的评估指标：
    Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.717
    Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.836
    Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.820
    Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.202
    Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.507
    Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.819
    Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.230
    Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.748
    Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.802
    Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.200
    Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.658
    Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.883
    06/11 16:40:03 - mmengine - INFO - bbox_mAP_copypaste: 0.717 0.836 0.820 0.202 0.507 0.819
  • 测试评估日志：20230611_163930.log

测试

  • 气球识别测试
    测试图片链接：
    测试图片1

  • 执行命令：

    python demo/image_demo.py \
data/tests/balloon/balloon_test.jpg \
rtmdet_tiny_1xb12-balloon.py \
--weights D:/github/mmdetection/outputs/balloon/best_coco_bbox_mAP_epoch_50.pth \
--pred-score-thr 0.42 \
--show

  • 预测结果：
    预测图1

特征可视化

  • backbone 特征可视化
  • 执行命令： 
    python demo/featmap_vis_demo.py \
data/tests/balloon/balloon_test.jpg \
../mmdetection/rtmdet_tiny_1xb12-balloon.py \
../mmdetection/outputs/balloon/best_coco_bbox_mAP_epoch_50.pth \
--target-layers backbone \
--channel-reduction squeeze_mean \
--show

backbone特征可视化

  • neck 特征可视化
  • 执行命令： 
    python demo/featmap_vis_demo.py \
data/tests/balloon/balloon_test.jpg \
../mmdetection/rtmdet_tiny_1xb12-balloon.py \
../mmdetection/outputs/balloon/best_coco_bbox_mAP_epoch_50.pth \
--target-layers neck \
--channel-reduction squeeze_mean \
--show

neck特征可视化

  • grad-cam boxAM可视化
  • 执行命令： 
    python demo/boxam_vis_demo.py \
data/tests/balloon/balloon_test.jpg \
../mmdetection/rtmdet_tiny_1xb12-balloon.py \
../mmdetection/outputs/balloon/best_coco_bbox_mAP_epoch_50.pth \
--target-layers neck.out_convs[2]  
--show

boxAM可视化